# US 12,216,998 B2 · Detecting online contextual evolution of linguistic terms > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,216,998 B2 - Document ID: US12216998B2 - Status: Granted - Kind: B2 - Application number: 18/377,690 - Filed: 2023-10-06 - Published: 2024-02-01 - Granted: 2025-02-04 - Earliest claimed: 2020-11-10 - Assignees: Discord Inc. - Applicants: Discord Inc. - Inventors: Michele Banko; Taylor Rhyne - Web page: https://patentgrep.com/patent/US12216998B2 ## CPC - G06F 40/279: Recognition of textual entities - G06F 40/284: Lexical analysis, e.g. tokenisation or collocates - G06F 40/30: Semantic analysis - H04L 51/52: for supporting social networking services - G06F 40/216: using statistical methods - G06N 3/02: Neural networks - G06N 20/00: Machine learning - H04L 51/18: Commands or executable codes - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract The present invention extends to methods, systems, and computer program products for detecting online contextual evolution of linguistic terms. Within messaging sources, some users may actively attempt to (relatively quickly) shift the meaning of a word or term. Some users may attempt to perjorate a word or term to have a more toxic meaning. Other users may attempt to reappropriate a word or term to have a less toxic or even a positive meaning. Aspects of the invention identify shifts in implied meanings of words and/or phrases over time. As such, emerging forms of harassment can be identified more quickly. Aspects of the invention can utilize users' behavioral histories as well as messaging structures to improve confidence when identifying term evolution. Machine learning algorithms can be configured to identify term evolution reducing workload on human moderators. ## Claims ### Claim 1 A computer-implemented method comprising: detecting an increase in use of a linguistic term at a messaging source; determining that a plurality of linguistic terms related to the linguistic term are disproportionally related to the increase in use of the linguistic term at the messaging source; determining a history of interaction using the linguistic term from a sending user to a receiving user at another messaging source; identifying a possible evolution of the linguistic term based on user reactions to the linguistic term at the messaging source, wherein the identifying the possible evolution considers the history of interaction; classifying one or more messages as including an evolved use of the linguistic term; and sending a message for confirmation of the evolved use by a moderator. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the messaging source has a history of using toxic language. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the messaging source has a history of using reappropriative language. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the identifying the possible evolution considers characteristics of the messaging source and user behavior histories of users interacting with messages containing the linguistic term, and wherein the characteristics and user behavioral histories are based on detecting that the sending user has used established language in addition to the linguistic term in other recent messaging interactions. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, further comprising placing the linguistic term on a watch list to monitor for pejorative use of the linguistic term. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, further comprising: computing a first sub-plurality of linguistic terms more related to the linguistic term based on a numerical value indicative of relatedness, the first sub-plurality of linguistic terms and the linguistic term included in the plurality of linguistic terms. ### Claim 7 A system comprising: one or more processors; and a non-transitory computer-readable medium storing instructions that, when executed by the one or more processors, cause the one or more processors to: detect an increase in use of a linguistic term at a messaging source; determine that a plurality of linguistic terms related to the linguistic term are disproportionally related to the increase in use of the linguistic term at the messaging source; determining a history of interaction using the linguistic term from a sending user to a receiving user at another messaging source; identify a possible evolution of the linguistic term based on user reactions to the linguistic term at the messaging source, wherein the identifying the possible evolution considers the history of interaction; classify one or more messages as including an evolved use of the linguistic term; and send a message for confirmation of the evolved use by a moderator. ### Claim 8 (depends on 7) The system of claim 7, wherein the messaging source is known to use toxic language. ### Claim 9 (depends on 7) The system of claim 7, wherein the messaging source has a history of using reappropriative language. ### Claim 10 (depends on 7) The system of claim 7, wherein the identifying the possible evolution considers characteristics of the messaging source and user behavior histories of users interacting with messages containing the linguistic term, and, wherein the characteristics and user behavioral histories are based on detecting that the sending user has used established language in addition to the linguistic term in other recent messaging interactions. ### Claim 11 (depends on 7) The system of claim 7, wherein the non-transitory computer-readable medium further causes the one or more processors to: place the linguistic term on a watch list to monitor for pejorative use of the linguistic term. ### Claim 12 (depends on 7) The system of claim 7, wherein the non-transitory computer-readable medium further causes the one or more processors to: compute a first sub-plurality of linguistic terms more related to the linguistic term based on a numerical value indicative of relatedness, the first sub-plurality of linguistic terms and the linguistic term included in the plurality of linguistic terms. ### Claim 13 (depends on 7) The system of claim 7, wherein the non-transitory computer-readable medium further causes the one or more processors to: receive positive or negative reaction to the linguistic term to shift a meaning associated with the evolved use. ### Claim 14 A non-transitory computer-readable storage medium comprising instructions stored thereon, the instructions effective to cause a system to: detect an increase in use of a linguistic term at a messaging source; determine that a plurality of linguistic terms related to the linguistic term are disproportionally related to the increase in use of the linguistic term at the messaging source; determine a history of interaction using the linguistic term from a sending user to a receiving user at another messaging source; identify a possible evolution of the linguistic term based on user reactions to the linguistic term at the messaging source, wherein the identifying the possible evolution considers the history of interaction; classify one or more messages as including an evolved use of the linguistic term; and send a message for confirmation of the evolved use by a moderator. ### Claim 15 (depends on 14) The non-transitory computer-readable storage medium of claim 14, wherein the messaging source is known to use toxic language. ### Claim 16 (depends on 14) The non-transitory computer-readable storage medium of claim 14, wherein the messaging source has a history of using reappropriative language. ### Claim 17 (depends on 14) The system of claim 14, wherein the identifying the possible evolution considers characteristics of the messaging source and user behavior histories of users interacting with messages containing the linguistic term, and, wherein the characteristics and user behavioral histories are based on detecting that the sending user has used established language in addition to the linguistic term in other recent messaging interactions.